感潮河段水体叶绿素三波段模型优化与应用

The Application of Three-band Inversion Model of Chlorophyll Concentration in the Saltwater Intrusion River

将地球植被的色素浓度的估算模型应用于珠江口感潮河段的水体叶绿素反演,并运用迭代优化方法优化了珠江口三波段叶绿素光谱模型。基于实测数据建立的叶绿素三波段反演模型,结果表明:叶绿素三波段模型线性关系直线斜率的变化与叶绿素a的均值浓度负相关,其第一波段随叶绿素a升高向短波方向移动;解释了81%的珠江河口叶绿素a变化,RMSE=1.4 mg/m^3,这个结果好于国内外同行的研究。EO-1 Hyperion的叶绿素制图结果表明,RMSE=3 mg/m^3(相对误差小于9.5%),叶绿素的空间分布也与地理趋势规律一致,为将来发射叶绿素制图传感器提供了依据。

The estimation model of vegetation pigment concentration is applied to the chlorophyll inversion in the tidal river section of the Pearl River Estuary and the three-band chlorophyll spectrum model of the Pearl River Estuary is optimized by iterative optimization method. The model based on measured data shows negative correlation between chlorophyll concentration and linear slope, and its first band moves to the short wave direction with chlorophyll increased. The three-band model explaines 81% chlorophyll of the Pearl River Estuary with RMSE=1.4 mg/m3 which is better than the research on domestic and foreign counterparts. The chlorophyll mapping results based on EO-1 Hyperion show that the spatial distribution of chlorophyll is consistent with the geographical trend with RMSE=3 mg/m3 （relative error was less than 9.5%）,which provides a basis for future sensors of chlorophyll mapping.